Detection of Cancer in Lymph Nodes by Spectral Imaging

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): This proposal presents data that demonstrate that lymph nodes can be analyzed for the presence of metastatic cancer cells by novel spectroscopic and mathematical methods. These methods use Infrared Spectral Imaging as the primary means for data acquisition, and multivariate statistical algorithms for the analysis of spectral data to produce a diagnosis. Specifically, this proposal will establish the scientific foundation for the detection of breast cancer cells in the axillary lymph nodes, excised during lumpectomies and mastectomies, in order to stage the spread of cancer. The goal of the proposed research is the development of an operating room-based instrument that will produce a diagnosis within 15 minutes after excision of a lymph node. The diagnosis will be presented in the form of a pseudo-color image that highlights the presence of breast cancer cells and/or secondary tumors. Thus, the milestones for this research are a) the development of rapid sample preparation methodology, b) the refinement of the imaging instrumentation, c) the construction of reliable databases and algorithms for the detection of metastatic cancer cells in lymph nodes, and d) the determination of sensitivity and specificity of the optical diagnostic method. The significance of the proposed detection method arises from the fact that many breast cancers are detected only after metastases have occurred. In such cases, the primary tumors have spread into adjacent lymph nodes, which need to be removed to stop the further proliferation of disease. The spread of metastatic cells determines further action, such as chemo- or radiation therapy. Thus, the accuracy and immediate availability of pathological results are of utmost importance. Presently, these results often are not available for hours or even days. The methodology described in this proposal is based on the observation from the PI's and other laboratories that cancerous tissue exhibits subtle infrared spectral differences that permit the distinction of normal and abnormal regions in biopsy tissue. These differences can be amplified by collecting thousands of individual spectra from microscopic regions of the tissue section (spectral imaging), and employing multivariate statistical methods of analysis. In this manner, images are obtained, in the form of pseudo-color maps, in which diseased areas are indicated. The methodology and technology to be developed in this research will readily be transferable to the analysis of lymph nodes affected with cancers other than breast cancer, once databases for other cancers are established. Thus, the proposed research represents a major improvement of present methods via novel optical and algorithmic diagnoses.